How to troubleshoot CIM hardware monitoring

KB ID:
1568
Product:
Veeam Management Pack for Microsoft System Center;Veeam Smart Plug-in (SPI) for VMware;Veeam ONE
Version:
5.7, 6.x, 7.x, 8.x
Published:
Last Modified:
2014-12-23

Challenge

The CIM API allows Veeam products to receive up-to-date information without requesting hardware refreshes, but this requires a separate connection, and if that connection cannot be made, the hardware data will be collected as it was in earlier versions, and request periodic hardware refreshes on the hosts.

Cause

There are the following ways in which the connection to the CIM API may fail:

  1. CIM Interaction is not enabled on the account used to connect to vCenter and ESX(i) hosts, or the account does not have direct access to the ESX(i) hosts. 
  2. The host name cannot be resolved from the collector or collectors.
  3. The CIM port is closed on the ESX/ESXi host or blocked by a firewall.
  4. The CIM server process may not be running.
  5. Updated hardware data are unavailable on the host for some reason.
  6. The time on the ESX host is different from the time in the domain.

Solution

The solutions below will allow you to troubleshoot and resolve the causes of CIM connection failures listed above:

  1. "CIM Interaction" should be enabled on the account used to monitor vCenter and the ESX/ESXi hosts. This is included in the Install Guide, but may have been overlooked, such as on an upgrade where the account had already been created.
  2. The host name must resolve from the collector system. The collector uses its existing connection to the vCenter server to obtain a name for the host, which is visible in the vSphere client when you select the host, go to the Configuration tab and select "DNS and Routing" on the left. If this name is unable to resolve from each collector that will monitor the host, or it contains a value such as "localhost" instead of the actual name and domain, the collector won't be able to connect to the host directly. If the name appears correct, but is unable to resolve anyway, you may need to update the DNS server or check for NAT firewalls on the network between the collector and host.
  3. The host must be accessible from the collector over port TCP/5989. The collector will communicate with the host using HTTPS over this port, so it must be accessible over the network through any firewalls and on the host itself, as described in VMware KB1013080. Note: if the port is closed, the failure to communicate with the host may interfere with the collector's attempts to retrieve data from other hosts.
  4. The CIM server must be running on the host to respond to the collector's requests for data. VMware has addressed problems in ESX 3.5 (VMware KB1009607) and ESX 4.0 prior to version 4.1 (VMware KB2011592).
  5. Check hardware data availability on VMware side. Open VIClient, select the affected host, go to "Hardware" tab, click "Reset sensors" and verify you do not have a warning saying "No new host data available. If from time to time you can see warnings like this, the subsequent troubleshooting lies on VMware side.User-added image
  6. The time on the ESX(i) hosts has to be synchronized with the domain, where Veeam is installed.

More Information

If you have any questions or problems, please contact Veeam Support at http://www.veeam.com/support

Please be aware that we’re making changes which will restrict access to product updates for users without an active contract.

OK

Rate the quality of this KB article: 
3.6 out of 5 based on 17 ratings

Couldn't find what you were looking for?

Below you can submit an idea for a new knowledge base article.

Report a typo on this page:

Please select a spelling error or a typo on this page with your mouse and press CTRL + Enter to report this mistake to us. Thank you!

Spelling error in text:

Submit